The Entropy Theory and Its Application in the Decision Tree
Baohua Yang · Computer Knowledge and Technology · 2006
Decision tree is a kind of predictive method for data to classify. In this paper the entropy theory and the typical algorithm of the ID3 are elaborated, and an example is proposed to build the decision tree. The attribute with the highest information gain is chosen to make the decision tree, namely the attribute with the highest discrimination as the current node, which is easy to transform the classified rules. The method needn't the domain knowledge to be examined and is simple to understand.